Final Report: Technical Assistance for the Gilson … · The Gilson Road (Sylvester) Superfund Site...

Final Report: Technical Assistance for the Gilson Road Superfund Site

Nashua, New Hampshire EPA Region 1

Solid Waste and EPA-542-R-09-012 Emergency Response September 2009 (5203P) www.epa.gov

Final Report: Technical Assistance for the Gilson Road Superfund Site

Nashua, New Hampshire EPA Region 1
http:www.epa.gov

Notice and Disclaimer

Work described herein was performed by GSI Environmental, Inc. for the U.S. Environmental Protection Agency (U.S. EPA) and has undergone technical review by EPA. Work conducted by GSI Environmental, Inc., including preparation of this report, was performed under EPA contract EP-W-07-037 to Environmental Management Support, Inc., Silver Spring. Maryland. Reference to any trade names, commercial products, process, or service does not constitute or imply endorsement, recommendation for use, or favoring by the U. S. EPA or any other agency of the United States Government. The views and opinions of the authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. For further information, contact:

Kirby Biggs Kathy Yager U.S. EPA/OSRTI U.S. EPA/OSRTI 703-299-3438 617-918-8362 [email protected] [email protected]
mailto:[email protected]:[email protected]

TABLE OF CONTENTS

EXECUTIVE SUMMARY ............................................................................................... i

1.0 INTRODUCTION................................................................................................. 1

1.1 Site Background...................................................................................................... 2

1.2 Remedial Activities................................................................................................. 3

1.3 Geology and Hydrology.......................................................................................... 4

1.4 Current Regulatory Status and Site Monitoring Objectives ................................... 5

2.0 MAROS EVALUATION...................................................................................... 6

2.1 Overburden Results................................................................................................. 6 2.1.1 COC Choice ................................................................................................ 6 2.1.2 Plume Stability............................................................................................ 7 2.1.3 Well Redundancy and Sufficiency............................................................ 10 2.1.4 Sampling Frequency ................................................................................. 11

2.2 Bedrock Aquifer.................................................................................................... 12 2.2.1 COC Choice .............................................................................................. 12 2.2.2 Plume Stability.......................................................................................... 12 2.2.3 Well Redundancy and Sufficiency............................................................ 13 2.2.4 Sampling Frequency ................................................................................. 14

2.3 Summary Results .................................................................................................. 14

3.0 CONCLUSIONS AND RECOMMENDATIONS............................................ 17

4.0 REFERENCES.................................................................................................... 20

TABLES........................................................................................................................... 21

Table 1: Gilson Road Monitoring Well Network Table 2: Priority Constituents, Screening Levels and Maximum Recent Concentrations Table 3: Aquifer Input Parameters Table 4: Trend Summary Results Overburden Aquifer Table 5: Trend Summary Results Bedrock Aquifer Table 6: Final Recommended Monitoring Network

FIGURES......................................................................................................................... 36

Figure 1: Gilson Road Site Monitoring Network Figure 2: Historic Conceptual Model Figure 3: Overburden Groundwater Arsenic and Benzene Average Concentrations

and Trend Results Figure 4: Combined Concentration Trends for Source and Tail

Figure 5: Spatial Uncertainty in Overburden Network Figure 6: Spatial Uncertainty in Final Recommended Overburden Network Figure 7: Bedrock Groundwater Arsenic and Benzene Average Concentrations and

Trend Results Figure 8: Final Recommended Monitoring Network

APPENDIX A: MAROS 2.2 METHODOLOGY....................................................... A-1

APPENDIX B: MAROS REPORTS ........................................................................... B-1

APPENDIX C: LIST OF ACRONYMS ..................................................................... C-1

EXECUTIVE SUMMARY

The following report reviews and provides recommendations for a long-term groundwater monitoring network for the Gilson Road (Sylvester) Superfund Site (Gilson Road). Extensive remedial actions have been successfully implemented at the site over the past 30 years, and the site is currently in a long-term operation and maintenance phase (O&M). The primary goal of developing an optimized groundwater monitoring strategy at the Gilson Road site is to create a dataset that fully supports site management decisions while minimizing expense and effort associated with long-term O&M.

The current groundwater monitoring network at the site has been evaluated using a formal qualitative approach as well as statistical tools found in the Monitoring and Remediation Optimization System software (MAROS). Recommendations are made for groundwater sampling frequency and location based on current hydrogeologic conditions as well as the long-term monitoring (LTM) goals for the site. The following report evaluates the monitoring system using analytical data collected from the site after cessation of the extraction remedy, including the time between 1999 and 2009. The report outlines recommendations based on a formal evaluation, but final determination of sampling locations and frequencies are to be decided by the overseeing regulatory agencies.

Site Groundwater Monitoring Goals and Objectives

Groundwater data at the Gilson Road site will be collected to address the following primary objectives:

Evaluate the risk to human health and the environment. Establish long term trends in contaminant levels to support future site

management decisions. Evaluate the effectiveness of the current remedial action (monitored natural

attenuation) in achieving risk reduction. Document changes to the area groundwater quality and geochemistry after

cessation of the groundwater extraction and treatment system. Ensure that contaminant concentrations above applicable screening levels are not

migrating horizontally and vertically to potential surface water receptors. Monitor groundwater concentrations at the boundaries of the groundwater

management zone (GMZ).

The goal of the long-term monitoring optimization (LTMO) analysis presented in this report is to review the current groundwater monitoring program and provide recommendations for improving the efficiency and accuracy of the network in supporting the site monitoring objectives listed above. Specifically, the LTMO process provides information on the site characterization, stability of the plume, sufficiency and redundancy of monitoring locations, and the appropriate frequency of sampling. The end product of the LTMO process at the Gilson Road site is a recommendation for specific sampling locations and frequencies that best address monitoring goals and support future

i

management and redevelopment decisions (see Figure 8 for the final network recommendations).

Results

Statistical and qualitative evaluations of the Gilson Road site analytical data have been conducted, and the following general conclusions have been developed based on the results of these analyses:

Historic remedial activities have diminished the size of the plume. The containment wall and groundwater extraction remedies have removed the majority of volatile organic contaminants (VOCs) from the overburden and bedrock aquifers. Arsenic is currently the contaminant of concern (COC) that exceeds cleanup standards at the most locations and by the highest amount.

Site characterization and conceptual model development are comprehensive and explain significant site details. No significant data gaps in site characterization were found. The current network is sufficient to support most site management decisions. However, due to the age of the site and the format and distribution of historic documents, relevant site data can be time-consuming to access.

Individual well trends and plume-wide trends indicate a stable to shrinking plume for all COCs in both the overburden and bedrock aquifers. Arsenic concentrations show strongly decreasing trends, particularly in the area downgradient of the slurry wall. Concentration trends for benzene, lead, and chlorobenzene are largely decreasing in both source (inside the slurry wall) and tail (outside the slurry wall) regions of the plume.

Chlorobenzene shows some variable trends in the overburden aquifer, outside of the slurry wall. Concentrations results for 2009 indicate chlorobenzene at well T-64-2 is just below the screening level; however, the concentrations show an overall increasing trend at this location. Chlorobenzene concentrations at HA-5-A have exceeded standards historically, but now show a decreasing trend. Nested wells at T-48 have some historic exceedances of the standard but now show a stable to decreasing concentration trend. Chlorobenzene concentrations at the downgradient boundary of the GMZ are below regulatory screening levels and show stable concentration trends.

Monitoring Well Redundancy/Sufficiency: Spatial analysis indicates networks in both aquifers can be reduced in the number of locations monitored. Overall, the aquifers show low variability and low uncertainty in concentrations.

Reduced Sampling Frequency: The statistical sampling frequency analysis along with a qualitative review indicated that a reduced sampling frequency (biennial) may be appropriate for many wells in the network.

ii

Recommendations

The following recommendations are made based on the results of the qualitative and quantitative review of data received, with findings summarized above and in Sections 3 and 4.

Plume Stability: Based on the results of the individual well trend and plume-wide stability analysis, the plumes in both the overburden and bedrock aquifers are stable to shrinking. Stable or shrinking plumes are candidates for reduction in monitoring effort.

Routine Monitoring Program: Several wells have been recommended for removal from the routine monitoring program for both the overburden and bedrock aquifers (see Table 6). For the overburden aquifer, 33 monitoring locations are recommended for retention in a routine monitoring program; 12 of these locations are recommended for biennial sampling with the remainder recommended for annual sampling. For the bedrock aquifer, 16 monitoring locations are recommended, with 3 at a biennial sampling frequency and the remainder recommended for annual sampling. Going forward, a consistent set of wells should be sampled at regular intervals to provide a dataset that supports plume-wide statistical evaluation of trends and plume-wide progress toward cleanup goals. A consistent dataset will provide a higher level of confidence in statistical results.

One additional bedrock monitoring well is recommended. While the spatial analysis indicates very low concentration uncertainty within the current network, there is currently no bedrock monitoring location at the northwestern boundary of the GMZ near HA-10 and HA-11. This area is downgradient from locations that exceed standards for arsenic and other COCs in the bedrock zone. A bedrock monitoring location in this area would provide information on concentrations at the edge of the institutional control (IC).

GMZ monitoring. One objective of the monitoring network is to confirm that groundwater outside of the GMZ meets quality standards. However, several wells that monitor the boundary of the current GMZ show concentrations above the background and some above the AGWS (e.g. HA-10-C for arsenic, T-54-3 for benzene and arsenic, T-60-3 for lead and arsenic). Technically, the GMZ must delineate the boundary between affected and unaffected groundwater. Based on results from the 2009 sampling, either the size of the GMZ must be adjusted or the requirements for groundwater attainment should be modified (e.g. calculating regional background concentrations for arsenic and lead). Additional monitoring locations may be required after expansion of the GMZ. Additional sampling locations may include the overburden downgradient from HA-10-C, and cross-gradient from HA-5-A and T-54-2. In addition to the bedrock well described above, another well may be necessary cross gradient from T-54-3.

Sampling Frequency: An annual sampling frequency is recommended for the majority of the monitoring locations and is recommended for locations in the source area and wells that monitor the downgradient area near Lyle Reed Brook. No locations are recommended for quarterly or semi-annual sampling. Biennial

iii

sampling is recommended for wells that delineate the GMZ or serve as point of compliance (POC) locations.

Data Management: Continue efforts to organize site data and transfer new and significant historical information to an electronic format to improve access to site data.

Chlorobenzene concentrations should be monitored and trends reviewed in the area immediately downgradient from the slurry wall in the overburden aquifer. Chlorobenzene concentrations at HA-5-A, T-64-2, and T-48-2, 3, and 4 should be carefully monitored for any increasing trends. Surface water in Lyle Reed Brook should be sampled downgradient from these locations in order to determine if concentrations exceed surface water quality standards.

Surface water and sediment monitoring: While surface water and sediment sampling locations were not evaluated in this report, the recommendation is to continue sampling the locations identified in the database on an annual basis along with groundwater locations.

Future reductions in monitoring effort may be possible if trends continue downward. After collection of a consistent dataset over a period of approximately 4 years, the network can be re-evaluated and reductions, particularly in sampling frequency may be appropriate.

iv

1.0 INTRODUCTION

The Gilson Road (Sylvester) Superfund Site is a National Priorities Listed (NPL) site near Nashua, New Hampshire in Region 1 of the U.S. Environmental Protection Agency (USEPA). The site comprises about 28 acres historically affected by the operation of an illegal waste disposal facility between the 1960s and 1979. Investigation and remediation activities began in the early 1980s, making the Gilson Road site one of the oldest sites to be managed under the Comprehensive Environmental Response and Liability Act (CERCLA or Superfund). Management of the site predated the 1986 Superfund Amendments and Reauthorization Act (SARA).

The Gilson Road site has undergone significant remedial activities over the past 30 years including isolation of a 20 acre parcel with a subterranean containment wall (slurry wall) and cap and installation of a groundwater pump and treat (P&T) system. Groundwater testing and monitoring began in 1981. Groundwater within the slurry wall was determined to have attained initial cleanup goals in 1995 and the active P&T remedy was terminated in 1996 (USEPA 2004). Groundwater monitoring efforts are currently underway to evaluate conditions after the cessation of the P&T remedy. Groundwater monitoring data will be used to evaluate whether monitored natural attenuation (MNA) is an appropriate long-term remedy for residual contamination. Therefore, current monitoring goals for the site include: 1) confirming that concentrations of constituents of concern (COCs) remain below relevant regulatory levels; 2) documenting changes to the groundwater quality and geochemistry after cessation of the P&T system; and 3) ensuring that COCs are not migrating horizontally and vertically to potential surface water receptors or beyond the boundaries of the groundwater management zone (GMZ). Groundwater data collected for the Gilson Road site may also be important in evaluating regional groundwater quality.

EPA Region 1 has requested GSI Environmental (GSI) under contract to EMS to review the Gilson Road site groundwater monitoring network and provide recommendations for improving the efficiency and accuracy of the network for supporting site management decisions. To this end, the following tasks have been performed:

Review monitoring objectives and current groundwater quality, and evaluate the ability of the monitoring network to achieve goals and objectives.

Evaluate individual well concentration trends over time, both within and outside of the slurry wall.

Evaluate overall plume stability through concentration trend and moment analysis.

Develop sampling location recommendations based on an analysis of spatial concentration uncertainty.

Develop sampling frequency recommendations based on both qualitative and quantitative statistical analysis results.

1

1.1 SITE BACKGROUND

The Gilson Road (Sylvester Site) is located near Nashua, New Hampshire about one-half mile east of the Nashua River, a tributary of the Merrimack River (see Figure 1). The site is bounded on the south by Gilson Road with low-density residential property to the south and west. Higher density residential property lies to the east and north of the site. The Four Hills Municipal Landfill is located to the northeast. Groundwater flow from the municipal landfill is in the direction of Lyle Reed Brook and the landfill may affect groundwater quality and geochemistry in this area. Lyle Reed Brook circles the site flowing northward from the west and bounding the site to the north. Lyle Reed Brook joins with Trout Brook northwest of the site, eventually discharging to the Nashua River to the northwest. The Nashua River joins the Merrimack River seven miles to the east of the site. The Merrimack is a water supply for the City of Lowell, Massachusetts.

The original source of contamination was a six-acre former sand and gravel borrow pit that was converted into an illegal solid waste disposal facility sometime in the 1960s by C & S Disposal Company. The disposal area was operated adjacent to the home of the owner, William Sylvester. The borrow pit was originally used to dispose of residential solid waste and demolition material; however, in the mid- 1970s the operator began accepting significant quantities of industrial hazardous wastes. Waste liquids and sludges containing VOCs, flammable solvents, heavy metal waste, and semivolatile organic compounds (SVOCs) (H&A 1994) were delivered to the site by tanker trucks and piped directly to the borrow pit or into subsurface leaching fields. Drums containing waste liquids and solids also were buried in the pit and stored on site.

A court order was issued in 1979 prohibiting further disposal of hazardous wastes at the site. In 1980, regulatory agencies acquired access to the property and removed 1,324 drums of primarily liquid BTEX (benzene, ethylbenzene, toluene, and xylenes) waste. Remedial investigation activities and an emergency response occurred between 1981 and 1982. Groundwater monitoring wells were installed in 1981, and a groundwater extraction system to contain affected groundwater was installed in 1982. A Record of Decision (ROD) was issued in July 1982 (USEPA 1982) requiring the construction of a slurry trench cutoff wall and surface cap isolating a 20-acre area. The slurry wall was constructed in December 1982 and several groundwater monitoring wells were installed during this time. (A list of current groundwater monitoring locations is provided in Table 1 and on Figure 1). A 1983 Supplemental ROD (SROD) (USEPA 1983) specified that a 300 gallon per minute (gpm) groundwater treatment plant be constructed to extract and treat affected groundwater from within the slurry wall. The 1983 SROD established cleanup goals, known as Alternative Concentration Limits (ACLs), for 16 constituents.

Because the remedial action was initiated before the widespread development of risk-based cleanup goals, the ACLs for the site were established at 90% of the original maximum concentrations of identified contaminants. ACLs were modified in a 2002 Explanation of Significant Differences (ESD) revising cleanup goals for 1,1-dichloroethane and 1,1,2-trichloroethane. ACLs apply to groundwater within the containment wall. No ACL for arsenic was specified in the SROD and analyses and state standards for 1,4-dioxane have only recently been developed.

2

In the intervening years, the state of New Hampshire has developed risk-based Ambient Groundwater Quality Standards (AGQS) and Ambient Water Quality Standards (AWQS) for surface water. The site currently has institutional controls (ICs) in place which incorporate a GMZ. Compliance at the boundary of the GMZ is based on the AGQS standards and AWQS apply to surface water in Lyle Reed Brook and the Nashua River. ACLs are the applicable standard within the containment wall. ACLs and AGQS values are shown in Table 2 for priority contaminants of concern (COCs) along with the most recent maximum concentrations in the plume. For the purpose of this report, ACLs are used to evaluate attainment of cleanup goals within the containment wall and to ensure an active remedy is not required in this area. AGQS apply outside the wall and to all compounds not specified in the SROD (e.g., arsenic, 1,4-dioxane).

1.2 REMEDIAL ACTIVITIES

At the time of the initial investigation, contaminated groundwater was estimated to be moving through the upper aquifer at a rate of 2 ft/day (Backers and Beljin 1996). The soil/bentonite slurry cutoff wall was constructed in September 1982 and consisted of a three-foot thick wall extending between 90 and 110 feet below ground surface (bgs) fully encompassing 20 acres (see Figure 1). A synthetic cover was installed over the site. The 300 gpm groundwater pump and treat (P&T) system was initiated in April 1986, becoming the first P&T system installed in the nation (USEPA 2004). Inorganic contaminants were removed from groundwater and disposed of in an onsite, lined landfill while volatile organic compounds (VOCs) were incinerated onsite. Following treatment, 250 gpm of effluent was discharged to trenches inside the slurry wall with 50 gpm discharged outside the slurry wall. Discharge within the slurry wall was intended to flush contaminants while discharge upgradient of the slurry wall was intended to raise the hydraulic head and facilitate groundwater migration from bedrock into the containment area. The remedial conceptual model from a 1989 report by Weston Solutions (Weston 1989) is illustrated on Figure 2. The groundwater extraction system was originally anticipated to run for three years.

The Gilson Road remedial system was reviewed in 1989 (Weston 1989), and an ESD was issued in 1990 (USEPA 1990). The 1990 ESD identified additional remedial measures including a soil vapor extraction system to address residual toluene and addition of six groundwater recovery wells to extend the capture zones to areas where contaminants had been redistributed by the trenching system. The ESD also stipulated than a Remedial Action Evaluation Study was to be conducted to evaluate the progress toward attaining ACLs. In 1994, the Remedial Action Evaluation Study (H&A 1994) concluded that the additional remedial measures had been successful and that groundwater was close to attaining ACLs within the containment wall. The groundwater P&T system was shut down in 1996 when the EPA determined that the cleanup goals set forth in the SROD had been attained. Between 1986 and 1996 the P&T system had pumped more than a billion gallons of water and removed more than 430,000 pounds of contaminants (USEPA 1997).

While several studies of the remedial system (Weston 1989) (H&A 1994) indicated that the slurry wall effectively prevented contaminant migration through the overburden, it

3

was known that contaminated groundwater was escaping through the bedrock flow zone. In a 1996 study, transport through the slurry wall was found to be minimal (Backers and Beljin 1996). However, groundwater migrating through the bedrock fractures beneath the cutoff wall was found to be substantial, with approximately 7,800 gal/day exiting the containment area (H&A 1994). Currently, one objective of the groundwater monitoring network is to document how leakage through and under the slurry wall affects surrounding ground and surface water.

ICs have been established at the site. A chain-link fence currently surrounds the 20-acre containment area and former treatment plant, and a GMZ has been established encompassing the containment area and downgradient locations around Lyle Reed Brook (see Figure 1). The current remedy at the site is monitored natural attenuation (MNA). Groundwater at the site has been monitored since 1997 with the objective of confirming that groundwater has attained standards within the slurry wall and that the plume is stable to decreasing outside of the slurry wall during the period since cessation of the active remedy.

While concentrations of VOCs were dramatically reduced as a result of the P&T system, concentrations of arsenic in site groundwater have remained fairly high. It is unclear how much of dissolved arsenic is a result of residual waste and how much may have been mobilized from endogenous rock by changes in site geochemistry. Regionally, groundwater from the Four Hills Landfill discharges to the Lyle Reed Brook area and data indicate elevated arsenic in this area as well. The Gilson Road monitoring network contributes to a regional network evaluating arsenic concentrations.

1.3 GEOLOGY AND HYDROLOGY

Regional geology consists of two principal subsurface zones: a stratified drift in the overburden (overburden aquifer) and a fractured biotite schist bedrock layer (bedrock aquifer). At the Gilson Road site, the overburden consists of anthropogenic fill, glacial outwash, and a glacial till. The sand and gravel borrow pit in the eastern portion of the site was filled with various types of refuse including construction/demolition debris during the 1960s. Test borings indicate that the fill ranges from 3 to 25 feet in depth and consists largely of coarse sand, gravel with bricks and wood fragments. The majority of the native overburden consists of glacial outwash, a coarse to fine sand with varying amounts of gravel and silt. The outwash ranges in thickness from 8 to 53 feet in depth, with thicker deposits to the south. Groundwater in the overburden aquifer is largely unconfined. A discontinuous layer of low-permeability glacial till separates the overburden from the bedrock and can be confining in some areas.

The bedrock surface varies with highs to the northeast and northwest, with a depression in the north central portion of the site. The bedrock elevation drops to the west of the site. The bedrock unit is moderately weathered and highly to moderately fractured. Fractures in the bedrock result in preferential groundwater flow paths. Groundwater in the bedrock aquifer is semi-confined in the secondary fractures. Based on the 1989 Weston report (Weston 1989), the two principal flow zones have similar transmissivity. A summary of

4

aquifer input parameters used in the analysis of overburden and bedrock groundwater is provided in Table 3.

The general groundwater flow direction is to the northwest toward the Nashua River. Groundwater flow in the overburden and bedrock are largely parallel. While the slurry wall contains overburden groundwater within the 20-acre enclosure, the hydraulic trend is vertically downward on the upgradient side of the northern slurry wall and vertically upward on the downgradient side (see Figure 2 for generalized conceptual model). Groundwater flowing downward under the wall then flows upward partially discharging to Lyle Reed Brook. Flow through the bedrock aquifer is toward the Nashua River where it discharges to the surface. With the current ICs in place, primary risk to environmental receptors focuses on human and ecological exposure pathways associated with discharge to Lyle Reed Brook.

Due to the age of many of the groundwater monitoring wells, accurate potentiometric surface measurements are difficult to obtain. Freezing and thawing of the ground can cause well casings to change position, resulting in inaccuracies in calculated depth to groundwater. Additionally, due to the age of the site, boring logs and as-built diagrams are not available for all wells and records of well installation are not uniform nor are they available in electronic format.

1.4 CURRENT REGULATORY STATUS AND SITE MONITORING OBJECTIVES

Since shutdown of the P&T system in 1996, groundwater monitoring has been conducted to confirm attainment of ACLs within the slurry wall and to monitor concentrations outside of the wall and in adjacent surface water as part of the MNA remedy. The site is currently in a verification stage to confirm that groundwater cleanup objectives will continue to be met under the current passive treatment scenario. In addition to groundwater monitoring, the slurry wall and surface cap and institutional controls are maintained.

Groundwater monitoring since 1999 has not been conducted on a regular schedule or with a regular group of wells. Additionally, results for 1,4-dioxane, used as an industrial solvent stabilizer during the 1970s, are limited to samples taken in 2009.

Based on the 2008 Draft Sampling and Analysis Plan (SAP) (NHDES 2008) for Gilson Road, the specific data quality objectives for the groundwater sampling program are to:

Evaluate the risk to human health and the environment. Establish long term trends in contaminant levels to support future site

management decisions. Evaluate the effectiveness of the remedial action in achieving risk reduction.

In order to address these objectives, the following monitoring location categories were used to design the network. Each groundwater monitoring location in the network was

5

evaluated qualitatively to determine how well it fulfilled one or more of the following functions:

Monitor possible exposure pathways such as discharge to surface water bodies (near Lyle Reed Brook);

Evaluate plume stability and possible migration of contaminants; Monitor the boundaries of the GMZ to ensure that concentrations do not exceed

regulatory limits outside of the IC; Monitor the historic source area inside of containment wall to confirm attenuation

of constituents and to anticipate future source strength;

Recommendations developed in the following report for the Gilson Road monitoring network are designed to address the objectives listed above. Results from both the qualitative evaluation and the statistical analyses contained in the MAROS software were reviewed to recommend optimized sampling locations and frequencies. Each well recommended for the final monitoring network (see Table 7) has been identified as addressing one or more of the monitoring objectives above.

2.0 MAROS EVALUATION

The MAROS 2.2 software was used to evaluate the LTM network at the Gilson Road site. MAROS is a collection of tools in one software package that is used to statistically evaluate groundwater monitoring programs. The tool includes models, statistics, heuristic rules, and empirical relationships to assist in optimizing a groundwater monitoring network system. Results generated from the software tool can be used to develop lines of evidence, which, in combination with professional judgment, can be used to inform regulatory decisions for safe and economical long-term monitoring of affected groundwater. A summary description of each tool and statistical method used in the analysis is provided in Appendix A of this report. For a detailed description of the structure of the software and further utilities, refer to the MAROS 2.2 Manual ((AFCEE 2004); http://www.gsi-net.com/software/MAROS_V2_2Manual.pdf) and Aziz et al., 2003 (Aziz, Newell, et al. 2003).

Groundwater data collected between 1999 and 2009, after total shutdown of the P&T system, were used for the majority of statistical analyses. Data from the overburden and bedrock aquifers were evaluated separately, despite the hydraulic connection and variability in vertical gradients between the two units. A summary of wells evaluated is presented in Table 1; regulatory screening levels are show in Table 2 and generalized aquifer input parameters for the MAROS software are presented in Table 3.

2.1 OVERBURDEN RESULTS

2.1.1 COC Choice

MAROS includes a short module that provides recommendations for prioritizing COCs plume-wide based on toxicity, prevalence, and mobility. A report showing results of the

6
http://www.gsi-net.com/software/MAROS_V2_2Manual.pdf

COC prioritization for the overburden aquifer is shown in Appendix B. Based on a comparison with AGQS screening levels (which are largely below the ACLs) arsenic, 1,4-dioxane and benzene are the only constituents that exceed standards on a plume-wide basis. Due to the low screening level (10 g/L), arsenic is the priority COC both inside and outside the containment wall. Benzene concentrations are well below ACLs within the slurry wall and have exceeded AGQS in the recent past in a limited area outside the wall at T-48-2/3/4, T64-2, and HA-5A/C.

The dataset for 1,4-dioxane is small, with results for only eight wells in the network from 2009. The AGQS for 1,4-dioxane is very low (3 g/L), so even low concentration detections can be problematic. Because of the limited dataset, trends for 1,4-dioxane as well as sampling locations and frequency could not be evaluated statistically. 1,4-Dioxane is highly mobile and detected at T-60-1, indicating that area impacts may originate from other sources such as the Four Hills Landfill. Additional data on the prevalence and distribution of 1,4-dioxane is needed.

Historically, chlorobenzene concentrations have exceeded AGQS standards in a limited area in the overburden aquifer outside the slurry wall (HA-5-A/C, T-48-2/3/4, and T64-2), but chlorobenzene does not exceed standards over a broad area, and recent (2009) samples indicate concentrations may be attenuating.

2.1.2 Plume Stability

Plume stability is an important concept in long-term site maintenance. A stable plume is one that is predictable under ambient conditions and requires less monitoring effort than plumes that are expanding or changing rapidly. Within the MAROS software, time-series concentration data and plume-wide trends are analyzed to develop a conclusion about plume stability.

Individual Well Trends

Data from 51 wells monitoring the overburden aquifer were evaluated. Summary statistics, including maximum detected concentrations (1999 2009), detection frequencies and concentration trends for arsenic, benzene, and chlorobenzene are shown in Table 4. Historic maximum concentrations for arsenic and benzene have been normalized by the AGQS and plotted on Figure 3 in order to provide an idea of the distribution of groundwater above the standards.

Individual well concentration trends were determined using the Mann-Kendall (MK) and linear regression (LR) methods for data collected between 1999 and 2009. A summary of trend results for the overburden aquifer is provided in the table below and in Table 4. Roughly one quarter of wells (12) sampled in the 2009 event have not been sampled more than three times in the previous ten years. A concentration trend cannot be calculated for locations with less than 4 sampling results. Detailed reports for MK trends are provided in Appendix B. Results of the individual well MK trends for arsenic and benzene are illustrated on Figure 3.

7

Overburden COC Total Number and Percentage of Wells for Each Trend Category Wells Non

Detect Decreasing/

Probably Decreasing

Stable Increasing/Probably

Increasing

No Trend N/A

Arsenic 51 1 (2%) 19 (37%) 13 (25%) 0 6 (12%) 12 (24%) Benzene 50 14 (28%) 8(16%) 13 (26%) 0 4 (8%) 11 (22%) Chlorobenzene 50 13 (26%) 6 (12%) 11 (22%) 3 (6%) 6 (12%) 11 (22%) Lead 51 7 (14%) 14 (27%) 10(19%) 0 8 (16%) 12 (24%)

Note: Number and percentage of total wells in each category shown. N/A = insufficient data to evaluate a trend.

For arsenic, the majority of well locations show decreasing to stable trends. This is true for the other constituents as well, with non-detect, decreasing and stable trends dominating results for wells with sufficient data to determine a trend. No locations show increasing or probably increasing trends for arsenic. Trend results indicate a shrinking arsenic plume, perhaps indicating that geochemical conditions conducive to arsenic mobility have reversed. Six wells have no trend for arsenic, indicating higher variability in the data.

While no increasing concentration trends were found for arsenic, benzene, lead, and vinyl chloride, three locations show an increasing trend for chlorobenzene. The three locations with increasing trends for chlorobenzene are T-13-3 and T-19-3 inside the slurry wall and T-64-2 outside of the wall (MK trend reports are in Appendix B). While concentrations are below AGQS at T-19-3, concentrations exceed AGQS at T-13-3 and are close to exceeding at T-64-2. An area of elevated chlorobenzene exists outside of the slurry wall around T-64-2, including HA-5-A and C, the T-48 nested wells and T-63-1. Concentrations of chlorobenzene are decreasing at location T-64-3, co-located with T-64-2 and screened approximately 30 feet deeper than T-64-2.

The MAROS software groups trend results from individual wells to determine a general trend for a specific area. For the overburden aquifer, arsenic trends within the slurry wall are generally stable while concentrations outside the wall show an overall decreasing trend. Figure 4 shows the combined MAROS trend results for priority constituents inside the slurry wall (Source Stability) and outside the wall (Tail Stability). For the five COCs evaluated, all show decreasing or probably decreasing concentration trends outside of the slurry wall and most show probably decreasing trends within the slurry wall. These results support the conclusion of a stable to shrinking plume.

Moment Analysis

Moment analysis was used to estimate the total dissolved mass (zeroth moment) and center of mass (first moment) for dissolved constituents for the full plume (both inside and outside the slurry wall) and for a limited number of wells outside of the slurry wall. Zeroth and first moments were found for annually consolidated data collected between 1999 and 2009, and an MK trend was determined for each. Due to variations in the number and identity of wells sampled during each event, annual consolidation of data

8

was necessary in order to calculate moments based on a more consistent set of wells. Results of the moment analysis of priority COCs for the full plume area are summarized in the table below.

Zeroth moments are rough estimates of total dissolved mass, assuming a constant porosity and uniform plume thickness across the site. Because of heterogeneities in the subsurface, the mass estimates are best used to calculate a trend of dissolved mass over time within the network rather than accurate calculations of total mass. The total dissolved mass estimate between 1999 and 2009 for arsenic is strongly decreasing. The total dissolved mass of benzene, chlorobenzene and lead were found to be stable. These results support the conclusion of a largely stable to decreasing plume.

Type of Moment Arsenic Benzene Chlorobenzene Lead

Zeroth Moment D S S S

First Moment NT I I PI

Second Moment X S NT S S

Second Moment Y S NT S S

Decreasing trend (D), Probably decreasing trend (PD), Stable (S), Probably Increasing trend (PI), and Increasing trend (I); (NT) No trend; (N/A) insufficient data to evaluate a trend.

The plume center of mass (first moment) was estimated for each year, and the distance of the center of mass from the source (assumed to be near T-33-1) was calculated. MK trends were evaluated for the distance of the center of mass from the source over time. No trend was seen in the center of mass for arsenic; however, benzene, chlorobenzene, and lead showed an increasing center of mass indicating that concentrations may have shifted downgradient over time. Because the total mass is stable, the results indicate that concentrations in the upgradient area are decreasing leaving more relative mass in the downgradient area of the plume. Centers of mass for all constituents evaluated are in the vicinity of well T-13. Centers of mass for arsenic and benzene are shown on Figure 3.

Second moments indicate the spatial distribution of mass between the center and the edge of the plume. Second moments in the X direction are metrics of the distribution of mass in the direction of groundwater flow, while those in the Y direction indicate the spread of mass orthogonal to groundwater flow. An increasing second moment would indicate an increase in mass at the edge of the plume relative to the center. For the overburden aquifer, most second moments are stable, with some variability seen in the second moments for benzene.

Moments calculated only for the plume outside of the slurry wall support the conclusion of stability. Annually consolidated data for 14 wells were evaluated for the priority constituents between 1999 and 2009. For arsenic outside the slurry wall, total mass was strongly decreasing and the center of mass was stable. For benzene, chlorobenzene and lead, total dissolved mass was stable. The center of mass for chlorobenzene was stable, and that for lead showed a probably decreasing trend. The center of mass for benzene showed not trend. No increasing trends were found for any of the constituents evaluated.

9

2.1.3 Well Redundancy and Sufficiency

Spatial analysis modules in MAROS recommend elimination of sampling locations that have little impact on the characterization of contaminant concentrations. Algorithms also identify areas within the monitoring network where additional wells may be needed. The spatial redundancy and sufficiency analysis for Gilson Road included a statistical analysis using data collected between 2006 and 2009 as well as a qualitative evaluation of well locations relative to monitoring objectives. For details on the MAROS redundancy and sufficiency analyses, see Appendix A or the MAROS Users Manual (AFCEE 2004).

Redundancy

A Delaunay mesh spatial analysis method was used to evaluate well redundancy for 54 wells in the overburden aquifer. The algorithm includes calculation of a slope factor (SF) that mathematically evaluates how well the concentration at a particular location can be estimated from the nearest neighbors. Because the analysis is for a two-dimensional slice of the aquifer, for each well nest, data from the screened interval with the highest concentration was used for the redundancy analysis. An average SF less than 0.30 was the criteria to identify a well that may provide redundant information and may be eligible for removal from the network. Average SFs for arsenic and the MAROS recommendation for elimination from the network are shown in Table 6. Results of the qualitative analysis were combined with the results of statistical analyses to make a final recommendation for inclusion of the well in the network.

The general results of the spatial redundancy analysis indicate overall low SFs and a moderate level of spatial redundancy. The results of the spatial redundancy analysis were considered along with the qualitative review of the function of the well in the network (also summarized in Table 6) in order to make the final recommendation. Some wells were recommended by the software for removal because they are located close together and have similar concentrations. However, for wells on opposite sides of the slurry wall (e.g. T-12-1/3 and HA-5-A/C), the monitoring objective of assessing contaminant passage through the slurry wall is served by close proximity of wells. For the most part, if the software recommended including one well in a nested group, the entire group was retained for vertical delineation. Of the 54 wells reviewed, 21 were recommended for removal from the program. For the overburden, 33 monitoring locations are recommended for inclusion in the program with varying sampling frequencies (see 2.1.4 Sampling Frequency).

Sufficiency

The well sufficiency module recommends potential locations for new wells in areas of high concentration uncertainty. The graphical results of the well sufficiency analysis for arsenic are shown on Figure 5. Like the redundancy analysis, well sufficiency is evaluated using SF. Areas between wells with higher SF, corresponding to higher concentration uncertainty, are candidates for new wells. For the Gilson Road overburden network, no areas of excess concentration uncertainty were found for the priority COCs within the current extent of the network. Overall, the plumes show very low spatial uncertainty, so no new wells are recommended.

10

In order to determine if removal of redundant wells causes excess spatial uncertainty, the sufficiency analysis was re-run with the final recommended well network. The results of the well sufficiency analysis of the final network for arsenic are shown on Figure 6. No excess concentration uncertainty resulted from removal of sampling locations from the program.

Because MAROS only evaluates well sufficiency within the current network, a qualitative review of the delineation of affected groundwater at the Gilson Road site was conducted. The boundary of the current GMZ is shown on Figure 1. Groundwater outside of the GMZ should be unaffected by site contaminants. The extent of the current GMZ presents some challenges for the plume outside of the slurry wall. Concentrations of arsenic at several sampling locations on the boundary of the GMZ are above the AGQS, such as HA-10-C, T-60-3, and T-54-2. Locations T-60-3 and T-62-2 exceed for lead. Groundwater at HA-5-A/C, very close to the cross-gradient boundary of the GMZ, exceeds standards for several constituents.

Currently, wells along the GMZ do not confirm that groundwater outside of the GMZ meets AGQS. For the inorganic constituents, arsenic and lead, no background concentrations are specified in the documents reviewed. If current lead and arsenic concentrations are a result of indigenous geochemical processes, then screening concentration levels may be adjusted and the GMZ does not need to be expanded. However, if the boundary of the GMZ changes, additional wells may be required for delineation. Specifically, additional wells below AGQS may be required downgradient from HA-10-C and T-60-1/3 and cross-gradient from HA-5-A/C and T-54-2.

2.1.4 Sampling Frequency

The recent sampling frequency and identity of wells at Gilson Road has not been consistent. Sampling has been roughly annual for the years between 2002 and 2006 with varying numbers of wells sampled (30 in 2003, 24 in 2004, 37 in 2005, and 29 in 2006). No samples were recorded in the years 2007 and 2008. A comprehensive sampling event was conducted in February and March 2009 where 47 wells were sampled.

Because of the uneven sampling interval, several wells in the network could not be evaluated for recent (2003 2009) rate of change and trends to determine an appropriate sampling interval. Wells with insufficient data within the recent sampling interval are assigned a default quarterly sampling frequency recommendation by the software in order to collect a sufficient amount of data. For wells with sufficient recent data, the MAROS results were considered along with other lines of evidence (see Table 6) to recommend a final sampling frequency. For wells with smaller datasets, sampling frequency was recommended based on the overall concentration trend and monitoring rationale for the well. Final recommendations are shown on Table 6 and on Figure 8.

Of the 33 wells recommended for the final network, 13 are recommended for biennial sampling (every two years). These can be sampled in alternate years (even and odd) or all every two years, depending on which is easier for contracting purposes. Wells recommended for biennial sampling function as point of compliance (POC) or GMZ

11

monitoring locations. 20 wells in the overburden aquifer are recommended for annual monitoring. These wells largely function as source monitoring locations (inside the slurry wall) and sentry wells that may indicate when concentrations in excess of standards may be migrating toward potential receptors. A summary of locations, frequencies, associated monitoring objectives, and suggested data analysis strategies is located in section 2.3.

2.2 BEDROCK AQUIFER

2.2.1 COC Choice

The results of the COC prioritization for the bedrock aquifer indicate that arsenic concentrations exceed the AGQC by the highest amount and at the largest number of monitoring locations. Lead and benzene also exceed AGQCs on a plume-wide basis, but exceedances are neither as high nor as widespread as those for arsenic. As in the overburden aquifer, 1,4-dioxane exceeds the AGQC, but analytical results for this constituent are so limited that it is difficult to determine if 1,4-dioxane is a long-term issue.

While chlorobenzene has been detected above AGQCs in the bedrock aquifer at three locations (T-12-4 inside the slurry wall and HA-5B and T-48-5), the distribution of chlorobenzene is limited.

2.2.2 Plume Stability

Individual Well Trend Analyses

MK and linear regression trend results for select constituents are shown on Table 5 and summarized below. Historic maximum concentrations for arsenic and benzene have been normalized by the AGQS and plotted on Figure 7 in order to provide an idea of the distribution of groundwater above the standards.

Overburden COC Total Number and Percentage of Wells for Each Trend Category Wells Non

Detect Decreasing/

Probably Decreasing

Stable Increasing/ Probably

Increasing

No Trend N/A

Arsenic 21 0 9 (43%) 5 (24%) 1 (5%) 1 (5%) 5 (24%) Benzene 21 7 (33%) 8 (38%) 5 (24%) 0 0 1 (5%) Chlorobenzene 21 8 (38%) 4 (19%) 3 (14%) 0 5 (24%) 1 (5%) Lead 21 3 (14%) 3 (14%) 8 (38%) 0 4 (19%) 3 (14%)

N/A = insufficient data to evaluate a trend.

As in the overburden aquifer, the majority of bedrock monitoring locations showed decreasing to stable trends for arsenic. In particular, wells located along and just outside of the northern section of the slurry wall show strongly decreasing trends (HA-5B, T-12-4, T-64-4, and T-48-5). The only location with an increasing trend for arsenic is the

12

upgradient bedrock location T-33-4, with an average concentration roughly twice that of the AGQC. T-33-4 is just inside the slurry wall. The adjacent location, T-32-4, is just outside the slurry wall and shows a probably decreasing trend for arsenic. These results may indicate some type of geochemical effect of the slurry wall on adsorption and desorption behavior of arsenic.

No increasing trends were found for benzene, chlorobenzene, or lead. Individual well trend results for bedrock indicate largely stable to decreasing concentrations for priority constituents and are consistent with reduced monitoring effort.

Moment Analysis

The results of the moment trend analyses are summarized below for the priority COCs. All bedrock wells were included in the analysis. As with the overburden, data were consolidated annually. Zeroth moments (estimates of total dissolved mass) for arsenic, benzene and chlorobenzene show decreasing to probably decreasing trends, indicating continued attenuation of COC concentrations after shut-down of the active remedy. Lead concentrations show no trend, due to higher variability in the data.

Type of Moment Arsenic Benzene Chlorobenzene Lead

Zeroth Moment PD D D NT

First Moment S PI S NT

Second Moment X NT I NT NT

Second Moment Y S I S S

Decreasing trend (D), Probably Decreasing trend (PD), Stable (S), Probably Increasing trend (PI), and Increasing trend (I); (NT) No Trend; (N/A) insufficient data to evaluate a trend.

Centers of mass over time for arsenic and benzene are shown on Figure 7. The first moment, or center of mass for arsenic is stable indicating that arsenic concentrations are decreasing uniformly across the network. The center of mass for benzene is probably increasing, however, the spatial variation in centers of mass over time is quite low relative to the size of the plume. All centers of mass for the bedrock network are close to well T-24-2/3.

2.2.3 Well Redundancy and Sufficiency

Redundancy

The well-redundancy analysis for the bedrock aquifer included a review of 22 wells. The bedrock aquifer was analyzed as one 2-dimensional slice. Average SFs calculated for arsenic and the MAROS recommendation for elimination from the network are shown in Table 6. Results of the qualitative analysis were combined with the results of statistical analyses to make a final recommendation for inclusion of each well in the network.

13

As in the overburden aquifer, the spatial analysis for the bedrock network indicates low variability and low uncertainty within the network. The general results of the spatial redundancy analysis indicate overall low SFs, with only 4 locations with SF for arsenic above 0.3 (T-62-3, T-33-4, T-99, and T-44-2). The results of the spatial redundancy analysis were considered along with the qualitative review of the function of the well in the network in order to make the final recommendation. Six locations were recommended for elimination from the network either due to MAROS recommendation (T-19-4 and T-100-2) or due to low SF and lack of sufficient monitoring rationale (T-38-2 and T-44-2) or insufficient recent data (T-29-3 and T-25-3). Sixteen bedrock locations are recommended for future monitoring. A summary of the recommended locations and monitoring rationales is provided in section 3.3.

Sufficiency

The well sufficiency analysis for the bedrock aquifer resulted in no recommendations for new monitoring locations. However, the algorithm is not designed to recommend locations outside of the current network. The extent of affected groundwater in the bedrock is not as well delineated as that in the overburden. Overburden locations HA-10 and HA-11 as well as T-63 are significantly downgradient of the 20-acre source and define the plume to the northwest at the boundary of the GMZ. The bedrock well T-99 monitors bedrock in the vicinity of the Nashua River; however, there are very few bedrock wells between locations HA-5-B, T-64-4, and T-48-5 and the Nashua River. In particular, the concentration of arsenic at the extent of the GMZ in bedrock is not known. A bedrock monitoring location in the area of HA-10 or HA-11 may provide important data for evaluating the regional geochemistry of arsenic and the extent of exceedance in the bedrock aquifer.

2.2.4 Sampling Frequency

The sampling history of the bedrock aquifer is similar to that of the overburden. The MAROS sampling frequency analysis was performed for locations with sufficient data (more than 4 recent sampling events). The final sampling frequency recommendation is based on both the quantitative rate of change estimates and a qualitative review based on the monitoring rationale of the location.

Of the 16 wells recommended for the final network, three are recommended for biennial sampling (every two years): T-32-4, G-42-2, and T-99. Wells recommended for biennial sampling function as POC or GMZ monitoring locations. Thirteen bedrock wells are recommended for annual sampling. A summary of locations, frequencies, and associated monitoring objectives as well as suggested data analysis strategies are located in section 3.3.

2.3 SUMMARY RESULTS

The final recommended monitoring network is summarized below and shown on Figure 8 and Table 6.

14

Monitoring locations have been recommended to address the monitoring objectives for delineating the plume, monitoring the GMZ boundary, assessing source attenuation and for monitoring the plume outside of the slurry wall for possible expansion. The recommended network contains 49 locations with an estimated average of 41 samples annually.

Overall results for the site indicate continued decreasing concentrations trends for COCs in most locations. In particular, arsenic concentrations appear to be strongly decreasing downgradient from the original source area. Statistical and qualitative results indicate a stable to shrinking plume in both bedrock and overburden aquifers during the time since cessation of the P&T remedy. Results are supportive of a reduction in monitoring effort for the site.

The table below summarizes the recommended monitoring network for the near future. As concentrations decrease with time, further reduction in monitoring effort may be appropriate.

15

Final Recommended Monitoring Network

Monitoring Objective Recommended Wells Number of

Wells

Recommended Sampling Frequency

Recommended Statistical Analysis

Overburden Bedrock

HA-10-A T-32-4 HA-10-B T-42-2 HA-10-C T-99 HA-11-A (possible new

Monitor GMZ Boundary or Point of Compliance or Upgradient Location

HA-11-B HA-11-C HA-13-B HA-14

well)

16 (+1)

Biennial

Detection Monitoring, Comparison Compare

detections with screening levels

HA-9-A T-32-3 T-42-1 T-62-2 T-98

Monitor GMZ Boundary or Point of Compliance

HA-4-B T-60-1 T-60-3

HA-4-A T-54-3

5 Annual Detection Monitoring, Compare detections with screening levels

HA-5-A HA-5-B HA-5-C HA-7-A HA-7-B T-48-5

Sentry/Plume Attenuation

T-48-2 T-48-3

T-62-3 T-64-4

12 Annual Statistical Trends;

95% UCL T-64-2 T-64-3

T-12-1 T-12-4 T-13-1 T-13-4 T-13-2 T-24-2 T-13-3 T-24-3

Source Attenuation T-19-1 T-24-1

T-33-4 T-8-3

16 Annual Statistical Trends; Comparison with

cleanup goals T-27-1 T-33-1 T-8-1 T-8-2

TOTAL Wells 49 TOTAL Samples Annually 41

Note: The recommended statistical trend analysis is Mann-Kendall, 95% UCL= upper confidence limit.

16

3.0 CONCLUSIONS AND RECOMMENDATIONS

Extensive remedial activities at the Gilson Road site have achieved groundwater cleanup standards set forth in the 1982 and 1983 RODs for the area within the containment wall. Extensive groundwater extraction and treatment has removed the overwhelming majority of VOC contaminants in groundwater. However, some residual contamination remains both within and outside of the slurry wall. While dissolved arsenic may not have been a major concern during the initial site investigation phase, arsenic is currently the major site COC in both the overburden and bedrock aquifers.

Arsenic concentrations have become more problematic both because regulatory screening levels have dropped from 50 g/L to 10 g/L (USEPA MCLs) and because changes in the geochemistry of the Gilson Road site may have enhanced desorption of arsenic from native sediments. Dissolved arsenic concentrations appear to result from a combination of historic waste disposal and geochemical conditions exacerbated by the installation and operation of the waste disposal and remedial systems. Part of the objective of the Gilson Road monitoring network is to evaluate regional arsenic geochemistry and the possible impact of both the Gilson Road site and the Four Hills Municipal Landfill on area ground and surface water.

Overall, arsenic concentrations are decreasing across the groundwater plume, both within the slurry wall and particularly downgradient of the slurry wall. A decreasing plume indicates that the monitoring effort may be reduced without loss of significant decision support metrics. Concentrations are also decreasing for benzene, chlorobenzene, and lead. Results for most other VOCs have dropped below detection limits. The center of mass of most of the constituent plumes is near well T-13 at the northern end of the containment area, where the majority of the monitoring effort is now centered.

Spatial redundancy and sufficiency analyses indicate very little spatial uncertainty in the plume and that the site has been well characterized by the number and location of the wells. Several locations are recommended for elimination from the routine monitoring program. No new locations are recommended for the overburden aquifer within the current network and only one possible downgradient POC/GMZ boundary well may be necessary for the bedrock aquifer. If the GMZ is expanded to encompass all groundwater currently above AGQC, additional overburden and bedrock wells may be required downgradient from HA-10 and cross-gradient from HA-5 and T-54.

Overall, statistical and qualitative analyses indicate that the sampling frequency can be reduced at most locations where concentrations are not changing rapidly. However, monitoring a consistent set of wells at regular intervals would provide a dataset that is easier to analyze and more robust to evaluate plume-wide trends and plume-wide progress toward cleanup goals.

Results and Recommendations

Result: Site characterization and conceptual model development are comprehensive and explain significant site details. No significant data gaps were

17

found. The current network is largely sufficient to support site management decisions. However, due to the age of the site and the format and distribution of historic documents, site data can be time-consuming to access.

Recommendation: Continue efforts to organize site data and transfer new and significant historical information to an electronic format. When possible, scan pages from historic site reports with boring logs, geologic cross sections, and remedial designs into electronic format.

Result: The sampling frequency and number and identity of wells sampled have been variable over the last 5 to 10 years.

Recommendation: Choose a specific set of wells and a regular sampling interval to institute over the next few years. A consistent set of wells sampled at regular intervals can provide important data for comparing site-wide trends over time and demonstrating site-wide compliance with cleanup goals. A consistent dataset will provide a higher level of confidence in statistical results.

Result: Historic remedial activities have diminished the size of the plume and removed the majority of VOCs. Arsenic is currently the contaminant of concern (COC) that exceeds cleanup standards at the most locations and by the highest amount in both the overburden and bedrock aquifers.

Recommendation: Optimize the groundwater monitoring network for arsenic and to a lesser extent, lead contamination. Continue to develop a regional conceptual model for arsenic fate and transport that includes possible contributions from changes in area geochemistry and the Four Hills Municipal Landfill.

Result: Individual well trends and plume-wide trends indicate a stable to shrinking plume for all COCs in both geologic formations.

Recommendation: Based on trend and stability analysis, reduction in monitoring effort is appropriate. With continued decreasing concentration trends, further reduction in monitoring effort, particularly in sampling frequency may be appropriate.

Result: Concentration trends for chlorobenzene are increasing at a limited number of locations in the overburden aquifer, including one location outside of the slurry wall.

Recommendation: Monitor chlorobenzene concentrations in the overburden area of HA-5, T-48 and T-64 nested locations outside of the slurry wall. Continue monitoring surface water in Lyle Reed Brook for chlorobenzene on an annual basis. If concentration trends continue to increase at T-64-2, consider monitoring the surface water semi-annually and outline possible triggers for installation of a contingent remedy for this location.

18

Result: Well redundancy analysis indicates that networks in both aquifers can be reduced in number. Overall, the aquifers show low variability in concentrations.

Recommendation: Several wells have been recommended for removal from the routine monitoring program for both the overburden and bedrock aquifers (see Table 6).

Result: Well sufficiency analysis indicates very low spatial uncertainty in the plumes and that no new monitoring locations are required within the current networks. However, there is currently no bedrock monitoring location at the northwestern boundary of the GMZ, downgradient from locations that exceed standards for arsenic. Also, not all overburden wells bounding the GMZ have concentrations below AGQS.

Recommendation: Install a new bedrock monitoring location in the vicinity of HA-10 or HA-11 to delineate arsenic impacts near the GMZ boundary. If the GMZ is modified, additional overburden wells may be necessary to delineate affected groundwater.

Result: Sampling frequency can be reduced at many locations due to the low rate of concentration change, the limited likelihood of plume migration and the reduced need for frequent management decisions.

Recommendation: Reduce the sampling frequency at many locations to biennial (every two years) and maintain annual sampling frequency within and just downgradient of the slurry wall for the next four years to confirm decreasing trends.

Result: Concentrations of COCs are decreasing across the site. ACLs already have been met within the containment area, and the site is progressing toward attainment of all cleanup goals.

Recommendation: Re-evaluate data needs in four years, and reduce both the number and frequency of sampling locations as appropriate for the designated land re-use.

Additional

While surface water and sediment sampling locations were not evaluated for this report, it is recommended that the locations indicated in the database be sampled annually, at roughly the same time as groundwater is sampled. Compliance with AWQC for arsenic, benzene and chlorobenzene should be confirmed along Lyle Reed Brook. Sampling for chlorobenzene downgradient from T-64-2 is particularly important as concentrations in this area are variable.

19

No recommendations have been made for a reduction in the analyte list for groundwater samples. The 2008 SAP indicates that some locations will only be sampled for arsenic and lead, and others for an expanded list of geochemical indicators. There is nothing in the analysis above that would counter-indicate this strategy and the approach appears reasonable.

4.0 REFERENCES

AFCEE (2004). Monitoring and Remediation Optimization Software User's Guide, Air Force Center for Environmental Excellence.

Aziz, J. A., C. J. Newell, et al. (2003). "MAROS: A Decision Support System for Optimizing Monitoring Plans." Ground Water 41(3): 355-367.

Backers, M. and M. Beljin (1996). Ground-Water Models of the Gilson Road Hazardous Waste Site. Ada, OK, US EPA RSKERL.

H&A (1994). Remedial Action Evaluation Study Gilson Road Superfund Site Nashua, New Hampshire. Concord, NH, New Hampshire Department of Environmental Services.

NHDES (2008). Sampling and Analysis Plan Gilson Road Superfund Site. Concord, NH, New Hampshire Department of Environmental Services.

USEPA (1982). Record of Decision: Sylvester. Washington D.C., US Environmental Protection Agency.

USEPA (1983). Record of Decision: Sylvester. Washington D.C., US Environmental Protection Agency.

USEPA (1990). Explanation of Significant Differences: Sylvester. Washington, D. C., US Environmental Protection Agency.

USEPA (1997). Sylvester/Gilson Road Superfund Site Verification of Attainment Phase. Boston, MA, US Environmental Protection Agency Region 1.

USEPA (2004). Five-Year Review Report: Third Five-Year Review Report for the Sylvester Superfund Site. Boston, US Environmental Protection Agency Region 1.

Weston (1989). Remedial Program Evaluation Gilson Road Site, Nashua, New Hampshire. Concord, New Hampshire, Roy F. Weston.

20

Groundwater Monitoring Network Optimization Gilson Road Superfund Site

Nashua, New Hampshire

TABLES

Table 1 Gilson Road Monitoring Well Network

Table 2 Priority Constituents, Screening Levels and Maximum Recent Concentrations

Table 3 Aquifer Input Parameters

Table 4 Trend Summary Results Overburden Aquifer

Table 5 Trend Summary Results Bedrock Aquifer

Table 6 Final Recommended Monitoring Network

21

Issued: 11-SEPT-2009 Page 1 of 2

TABLE 1 GILSON ROAD MONITORING WELL NETWORK

Long-Term Monitoring Optimization Gilson Road Superfund Site, Nashua, New Hampshire

Well Name Interior or Exterior of Slurry Wall

Screened Interval

(FT below TOC)

Total Depth (FT below

TOC)

Minimum Sample Date

Maximum Sample Date

Number of Samples

1999 - 2009 Priority Constituent Above or Below AGQS

Overburden Locations HA-10-A Exterior 12/1/1999 2/27/2009 10 ARSENIC Below HA-10-B Exterior 12/1/1999 2/27/2009 11 ARSENIC Below HA-10-C Exterior 11/29/1999 2/27/2009 11 ARSENIC Above HA-11-A Exterior 12/1/1999 3/4/2009 11 LEAD Below HA-11-B Exterior 12/1/1999 3/4/2009 11 ARSENIC Below HA-11-C Exterior 11/29/1999 3/16/2009 11 ARSENIC Below HA-12-A Exterior 6/2/2005 3/10/2009 3 ARSENIC Below HA-12-B Exterior 6/2/2005 3/10/2009 3 None Below HA-12-C Exterior 6/2/2005 3/10/2009 3 None Below HA-13-B Exterior 12/2/1999 3/17/2009 12 ARSENIC Above HA-14 Exterior 12/2/1999 3/16/2009 10 ARSENIC Above HA-4-B Exterior 24 29 7/25/2003 3/9/2009 6 LEAD Below HA-5-A Exterior 50 55 12/3/1999 3/5/2009 14 ARSENIC Above HA-5-C Exterior 20 25 12/3/1999 3/5/2009 13 ARSENIC Above HA-7-B Exterior 5 15 12/3/1999 3/10/2009 9 ARSENIC Above HA-9-A Exterior 96 112.2 12/2/1999 3/16/2009 11 ARSENIC Below T-100-1 Exterior 12/1/2000 3/17/2009 11 LEAD Above T-12-1 Interior 19 29 12/9/1999 3/6/2009 14 ARSENIC Above T-12-3 Interior 55 12/9/1999 4/25/2002 7 ARSENIC Above T-13-1 Interior 15 25 12/10/1999 3/6/2009 16 ARSENIC Above T-13-2 Interior 39 12/10/1999 3/12/2009 14 ARSENIC Above T-13-3 Interior 55 12/10/1999 3/12/2009 15 ARSENIC Above T-18-1 Interior 10 20 12/10/1999 3/6/2009 4 ARSENIC Above T-18-2 Interior 40 12/10/1999 4/12/2000 2 ARSENIC Above T-18-3 Interior 65 12/10/1999 4/12/2000 2 ARSENIC Above T-19-1 Interior 5 15 12/10/1999 3/6/2009 11 ARSENIC Above T-19-3 Interior 50 12/10/1999 3/12/2009 8 LEAD Above T-24-1 Interior 0 12/8/1999 3/6/2009 16 ARSENIC Above T-25-1 Interior 28 38 12/8/1999 3/6/2009 9 ARSENIC Above T-25-2 Interior 48 12/8/1999 4/25/2002 7 ARSENIC Above T-27-1 Interior 12/7/1999 3/10/2009 8 ARSENIC Above T-32-3 Exterior 65 12/6/1999 3/12/2009 11 ARSENIC Below T-33-1 Interior 17.5 27.5 12/7/1999 3/9/2009 15 ARSENIC Above

T-33-2 Interior 45 12/7/1999 6/9/2005 10 METHYLENE CHLORIDE Above

T-34-1 Interior 25 35 12/6/1999 3/9/2009 12 ARSENIC Above T-42-1 Exterior 17 27 7/28/2003 3/9/2009 6 ARSENIC Below T-44-1 Exterior 15 25 7/16/2003 3/6/2009 6 ARSENIC Above T-47 Exterior 1.2 6.2 12/6/1999 3/9/2009 8 ARSENIC Above T-48-2 Exterior 32 37 12/6/1999 3/10/2009 15 ARSENIC Above T-48-3 Exterior 62 67 12/6/1999 3/10/2009 16 ARSENIC Above T-48-4 Exterior 85.5 12/6/1999 3/12/2009 8 ARSENIC Above T-54-2 Exterior 43 7/16/2003 3/13/2009 6 ARSENIC Above T-58 Exterior 12/2/1999 3/16/2009 7 CHLOROFORM Below T-60-1 Exterior 3 8 12/2/1999 3/9/2009 13 1,4-DIOXANE Above T-60-3 Exterior 36.5 38 12/2/1999 3/17/2009 13 LEAD Above T-61 Exterior 12/2/1999 3/13/2009 13 LEAD Above T-62-2 Exterior 28.5 30 10/18/2000 3/13/2009 9 LEAD Above T-63-1 Exterior 13.3 18.3 12/2/1999 3/4/2009 10 ARSENIC Above T-64-2 Exterior 33.8 38.8 12/2/1999 3/4/2009 8 ARSENIC Above T-64-3 Exterior 64.8 69.8 12/2/1999 4/24/2002 7 ARSENIC Above T-8-1 Interior 10 20 12/9/1999 3/6/2009 7 ARSENIC Above T-8-2 Interior 38.2 12/9/1999 3/13/2009 7 ARSENIC Above T-97 Exterior 7/23/2003 6/9/2005 4 None Below T-98 Exterior 11.5 14 7/23/2003 3/9/2009 6 CHLOROFORM Below See notes end of table

22


TABLE 1 GILSON ROAD MONITORING WELL NETWORK

Long-Term Monitoring Optimization Gilson Road Superfund Site, Nashua, New Hampshire

Well Name Interior or Exterior of Slurry Wall

Screened Interval

(FT below TOC)

Total Depth (FT below

TOC)

Minimum Sample Date

Maximum Sample Date

Number of Samples

1999 - 2009 Priority Constituent Above or Below AGQS

Bedrock Locations HA-4-A Exterior 49.3 7/25/2003 3/16/2009 6 ARSENIC Above HA-5-B Exterior 77.2 12/3/1999 3/5/2009 18 ARSENIC Above HA-7-A Exterior 12/3/1999 3/10/2009 14 ARSENIC Above T-100-2 Exterior 12/1/2000 3/17/2009 11 ARSENIC Above T-12-4 Interior 70 12/9/1999 3/12/2009 17 ARSENIC Above T-13-4 Interior 66.7 12/10/1999 3/12/2009 9 ARSENIC Above T-19-4 Interior 64 12/10/1999 3/12/2009 13 ARSENIC Above T-24-2 Interior 60 12/8/1999 3/13/2009 8 ARSENIC Above T-24-3 Interior 95.2 12/8/1999 3/13/2009 8 ARSENIC Above T-25-3 Interior 62.6 12/8/1999 4/25/2002 7 ARSENIC Above T-29-3 Interior 79.75 12/7/1999 3/13/2009 9 ARSENIC Above T-32-4 Exterior 89 12/6/1999 3/12/2009 13 LEAD Above T-33-4 Interior 81 12/7/1999 7/18/2006 13 ARSENIC Above T-38-2 Exterior 47.9 12/7/1999 3/17/2009 9 ARSENIC Above T-42-2 Exterior 40.33 7/28/2003 6/9/2005 4 ARSENIC Below T-44-2 Exterior 38.4 7/16/2003 3/13/2009 6 LEAD Above T-48-5 Exterior 97 12/6/1999 5/5/2004 10 ARSENIC Above T-54-3 Exterior 60 7/16/2003 3/13/2009 6 ARSENIC Above T-62-3 Exterior 44 45.5 10/18/2000 3/13/2009 9 LEAD Above T-64-4 Exterior 97.5 12/2/1999 3/17/2009 10 ARSENIC Above T-8-3 Interior 60.9 12/9/1999 3/13/2009 16 ARSENIC Above T-99 Exterior 26.5 29 7/23/2003 3/9/2009 6 ARSENIC Below

Notes: 1. Well screened intervals, locations and sample history from the Weston database, 2009. 2. AGQS = Ambient Groundwater Quality Standard for New Hampshire (see Table 2). 3. Priority constituent determined by normalizing historic maximum concentrations by the AGQS.

The constituent with the highest concentration to screening level ratio is the priority COC for the well. 4. Above = Locations with maximum concentrations of any constituent over the AGQS data 1999 - 2009.

23

Issu

ed 1

1-S

EP

T-20

09

Pag

e 1

of 1

TAB

LE 2

PR

IOR

ITY

CO

NST

ITU

ENTS

, SC

REE

NIN

G L

EVEL

S A

ND

MA

XIM

UM

REC

ENT

CO

NC

ENTR

ATI

ON

S

LON

G-T

ERM

MO

NIT

OR

ING

OPT

IMIZ

ATI

ON

G

ilson

Roa

d Su

perf

und

Site

, Nas

hua,

New

Ham

pshi

re

24

Con

stitu

ent

AC

L [u

g/L]

A

GQ

S [u

g/L]

M

CL

[ug/

L]

Max

imum

C

once

ntra

tion

Sprin

g 20

09

1,1,

1-Tr

ichl

oroe

than

e (T

CA

) 1,

1,2-

Tric

hlor

oeth

ane

1,1-

Dic

hlor

oeth

ane

1,2-

Dic

hlor

oeth

ane

Ben

zene

C

hlor

oben

zene

C

hlor

ofor

m

Lead

M

ethy

l Eth

yl K

eton

e (2

-But

anon

e)

Met

hyl M

etha

cryl

ate

Met

hyle

ne C

hlor

ide

Phe

nols

S

elen

ium

Te

trach

loro

ethe

ne (P

CE

) To

luen

e Tr

ichl

oroe

then

e (T

CE

) V

inyl

Chl

orid

e

200 3 81

1800

34

0 11

0 15

05

8000

35

0 12

250

400

2.6

57

2900

15

00

95

200 5 81

15

5 100

70

15

4000

5 40

00

50

5 10

00

5 2

200 5 5 5 100

15

5 50

5 10

00

5 2

50

27

42

316

4.8

84.2

17

1890

10

0 5.

3

Ars

enic

1,

4-D

ioxa

ne

10

3 10

10

10

82

Not

es:

1. A

CL

= A

ltern

ate

Con

cent

ratio

n Li

mits

from

198

2 R

OD

mod

ified

in 2

002

ES

D (E

PA

, 200

2).

2. A

GQ

S =

Sta

te o

f New

Ham

pshi

re A

mbi

ent G

roun

dwat

er Q

ualit

y S

tand

ards

. 3.

MC

L =

US

EP

A M

axim

um C

onta

min

ant L

evel

. 4.

Ars

enic

and

1,4

-Dio

xane

wer

e no

t ide

ntifi

ed in

the

RO

D a

s co

ntam

inan

ts

of c

once

rn, b

ut e

xist

at t

he s

ite a

bove

NH

scr

eeni

ng le

vels

. 5.

Max

imum

con

cent

ratio

ns fr

om s

ite-w

ide

data

col

lect

ed in

Feb

rura

y to

Mar

ch 2

009.

Issued 11-SEPT-2009 Page 1 of 1

TABLE 3 AQUIFER INPUT PARAMETERS

LONG-TERM MONITORING OPTIMIZATION Gilson Road Superfund Site

Parameter Value Units Porosity(n) 0.3 Seepage velocity 365 ft/yr Plume Thickness 20 ft Plume Length 1500 ft Plume Width 650 ft Distance to Receptors (Lyle Reed Brook) GWFluctuations

SourceTreatment Contaminant Type NAPLPresent

1500 Yes

Cap and slurry wall/historic pump and treat

Chlorinated solvents/metals No

ft --

------

Groundwater flow direction (N/NW) N/NW 135 Source Location near Well T-33-1 --Source X-Coordinate 1023045 ft Source Y-Coordinate 79861.84 ft Coordinate System NAD 83 SP New Hampshire Non-detect values Set to lowest detection limit

Notes: 1. Aquifer data from Weston, 1989 and Haley and Aldrich, 1994. 2. Data above were used for both overburden and bedrock aquifers.

25


TABLE 4 TREND SUMMARY RESULTS OVERBURDEN AQUIFER

LONG-TERM MONITORING OPTIMIZATION Gilson Road Superfund Site, Nashua, New Hampshire

WellName Number of Samples

Number of Detects

Percent Detection

Maximum Result 1999 -

2009 [ug/L]

Max Result Above

Standard?

Average Result 1999 -

2009 [ug/L]

Average Result Above

Standard?

Mann-Kendall Trend

Linear Regression

Trend ARSENIC HA-10-A HA-10-B HA-10-C HA-11-A HA-11-B

9 9 9 9 9

7 1 9 8 1

78% 11%

100% 89% 11%

2.2 1.1 50.3 3.8 1.4

No No Yes No No

1.38 1.01

38.70 2.58 1.04

No No Yes No No

S S

NT S S

S D

NT S S

HA-11-C HA-12-A HA-12-B HA-12-C HA-13-B

9 1 1 1

10

5 1 0 0 6

56% 100%

0% 0% 60%

7.1 6.4 ND ND 10.9

No No ND ND Yes

2.21 6.40 ND ND 3.19

No No ND ND No

D N/A N/A N/A S

D N/A N/A N/A NT

HA-14 HA-4-B HA-5-A HA-5-C HA-7-B

1 1

11 10 8

1 0

11 10 8

100% 0%

100% 100% 100%

14 ND 796 580 198

Yes ND Yes Yes Yes

14 ND 675 542 66

Yes ND Yes Yes Yes

N/A N/A D D

NT

N/A N/A D D

NT HA-9-A T-100-1 T-12-1 T-12-3 T-13-1

1 9

10 6

11

1 3

10 6

11

100% 33%

100% 100% 100%

2.2 18.3 496 889 399

No Yes Yes Yes Yes

2.15 4.29 378 786 218

No No Yes Yes Yes

N/A NT D

NT D

N/A NT D

NT D

T-13-2 T-13-3 T-18-1 T-19-1 T-19-3

11 11 1 8 7

11 11 1 8 7

100% 100% 100% 100% 100%

633 1400 395 114 4.2

Yes Yes Yes Yes No

572 965 395 52.7 2.66

Yes Yes Yes Yes No

S D

N/A D

NT

S D

N/A D PI

T-24-1 T-25-1 T-25-2 T-27-1 T-32-3 T-33-1

11 7 6 1 9

10

11 7 6 1 4

10

100% 100% 100% 100% 44%

100%

605.0 759.0 805.0 136.0

3.6 705

Yes Yes Yes Yes No Yes

524 604 682 136 1.66 149

Yes Yes Yes Yes No Yes

D S

NT N/A D D

D S PI

N/A D D

T-33-2 T-34-1 T-42-1 T-44-1 T-47

8 10 1 1 7

3 10 1 1 7

38% 100% 100% 100% 100%

1.5 2120 1.2 28.9 627

No Yes No Yes Yes

1.11 347 1.20 29 342

No Yes No Yes Yes

S D

N/A N/A S

S D

N/A N/A D

T-48-2 T-48-3 T-48-4 T-54-2 T-58

11 11 7 1 6

11 11 7 1 0

100% 100% 100% 100%

0%

693 703 685 18.1 ND

Yes Yes Yes Yes ND

550 517 566 18 ND

Yes Yes Yes Yes ND

D PD PD N/A ND

D D D

N/A ND

T-60-1 T-60-3 T-61 T-62-2 T-63-1

11 11 11 7 8

8 11 10 3 8

73% 100% 91% 43%

100%

3.5 30.4 9.7 1.9

1870

No Yes No No Yes

1.79 11

2.09 1.23 963

No Yes No No Yes

D D D S D

D S

NT PD D

T-64-2 T-64-3 T-8-1 T-8-2 T-98

7 6 6 6 1

7 6 6 6 0

100% 100% 100% 100%

0%

1050 843 455 401 ND

Yes Yes Yes Yes ND

852 679 368 251 ND

Yes Yes Yes Yes ND

S D S S

N/A

S D S S

N/A See notes end of table

26





Number of Detects

Percent Detection


2009 [ug/L]

Max Result Above

Standard?


2009 [ug/L]


Standard?

Mann-Kendall Trend

Linear Regression

Trend Benzene HA-10-A HA-10-B HA-10-C HA-11-A HA-11-B

7 7 7 7 7

0 0 0 0 0

0% 0% 0% 0% 0%

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND


7 1 1 1 9

0 0 0 0 0

0% 0% 0% 0% 0%

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND N/A N/A N/A ND

ND N/A N/A N/A ND


1 1

11 11 4

0 0

11 11 0

0% 0%

100% 100%

0%

ND ND 11 7.3 ND

ND ND Yes Yes ND

ND ND 7 5

ND

ND ND Yes Yes ND

N/A N/A D S

ND

N/A N/A D D

ND HA-9-A T-100-1 T-12-1 T-12-3 T-13-1

1 9

11 6

11

0 0 9 6 4

0% 0% 82%

100% 36%

ND ND 6.1 9.8 5.2

ND ND Yes Yes Yes

ND ND 3.42

6 3

ND ND No Yes No

N/A ND S

PD PD

N/A ND S D D

T-13-2 T-13-3 T-19-1 T-19-3 T-24-1

11 11 4 6

11

11 11 0 2

11

100% 100%

0% 33%

100%

30 36 ND 7.4 26

Yes Yes ND Yes Yes

15 21 ND 3.17

18.80

Yes Yes ND No Yes

D D

ND PD S

D D

ND D S

T-25-1 T-25-2 T-27-1 T-32-3 T-33-1 T-33-2

7 6 1 9

11 4

7 6 1 0 6 1

100% 100% 100%

0% 55% 25%

51.0 23.0 5.7 ND 8.2 2.2

Yes Yes Yes ND Yes No

28 13 5.7 ND 3.49

2

Yes Yes Yes ND No No

S S

N/A ND D

NT

D D

N/A ND D

NT T-34-1 T-42-1 T-44-1 T-47 T-48-2

9 1 1 5

11

1 0 0 0

10

11% 0% 0% 0% 91%

27 ND ND ND 7.8

Yes ND ND ND Yes

4.78 ND ND ND 5

No ND ND ND Yes

NT N/A N/A ND S

NT N/A N/A ND D

T-48-3 T-48-4 T-54-2 T-58 T-60-1

11 7 1 5 9

11 7 0 0 2

100% 100%

0% 0% 22%

8.9 8.4 ND ND 2.2

Yes Yes ND ND No

6 6

ND ND 2.03

Yes Yes ND ND No

S NT N/A ND S

D D

N/A ND S

T-60-3 T-61 T-62-2 T-63-1 T-64-2

9 9 7 8 7

1 2 0 7 7

11% 22% 0% 88%

100%

2.2 2.9 ND 8.7 13

No No ND Yes Yes

2.02 2

ND 4.76

6

No No ND No Yes

S S

ND PD NT

D S

ND PD S

T-64-3 T-8-1 T-8-2 T-98

5 7 7 1

1 6 7 0

20% 86%

100% 0%

4.4 42 55 ND

No Yes Yes ND

2 25

28.1 ND

No Yes Yes ND

S S S

N/A

PD NT S

N/A See notes end of table

27





Number of Detects

Percent Detection


2009 [ug/L]

Max Result Above

Standard?


2009 [ug/L]


Standard?

Mann-Kendall Trend

Linear Regression

Trend Chlorobenzene HA-10-A HA-10-B HA-10-C HA-11-A HA-11-B

7 7 7 7 7

0 0 0 0 0

0% 0% 0% 0% 0%

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND


7 1 1 1 9

0 0 0 0 0

0% 0% 0% 0% 0%

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND ND ND ND ND

ND N/A N/A N/A ND

ND N/A N/A N/A ND


1 1

11 11 4

0 0

11 11 2

0% 0%

100% 100% 50%

ND ND 160 130

7

ND ND Yes Yes No

ND ND 130 104

4

ND ND Yes Yes No

N/A N/A D S S

N/A N/A S S S

HA-9-A T-100-1 T-12-1 T-12-3 T-13-1

1 9

11 6

11

0 0

11 6

11

0% 0%

100% 100% 100%

ND ND 118 90 87

ND ND Yes No No

ND ND 72.8 77 49

ND ND No No No

N/A ND D

NT NT

N/A ND S PI I

T-13-2 T-13-3 T-19-1 T-19-3 T-24-1

11 11 5 6

11

11 11 5 6

11

100% 100% 100% 100% 100%

123 150 28 36 470

Yes Yes No No Yes

68 67

14.6 13.1

219.00

No No No No Yes

NT I S I

PD

NT I S I S

T-25-1 T-25-2 T-27-1 T-32-3 T-33-1 T-33-2

7 6 1 9

11 4

7 6 1 0

11 0

100% 100% 100%

0% 100%

Date post:	14-Oct-2018
Category:	Documents
Upload:	hanhu
View:	232 times
Download:	0 times

Final Report: Technical Assistance for the Gilson … · The Gilson Road (Sylvester) Superfund Site...

Documents